Abstract Recently, gene expression profiling has identified molecular subtypes that classify invasive breast cancers into distinct categories that vary in their clinical behavior and response to treatment. These subtypes highlight the many possible biologically and clinically distinct types of breast cancer. With such heterogeneity within breast cancer, we might expect that risk factors influence specific subtypes of breast cancer through different etiologic pathways. Mammographic density (MD), the proportion of the white or dense regions on a mammogram, has been shown to be one of the strongest and most consistent risk factors for breast cancer. We propose a large collaboration of four established cohort studies [Nurses' Health Study (I and II) Blood Subcohorts, Mayo Mammography Health Study and San Francisco Mammography Registry/UCSF SPORE] with similar methods for ascertaining risk factors, MD, breast cancer and breast cancer subtype information, to examine the association of MD with subtypes of invasive breast cancer. Specifically, we propose Aim 1) To ascertain and combine clinical risk factor, MD and breast cancer information to create a large nested case-control study of ~3300 women with invasive breast cancer and ~5700 matched controls, Aim 2) To characterize breast cancer subtypes for all invasive breast cancers using information from pathology reports, cancer registries and pathologic review. We propose to classify cancers according to hormone receptors (estrogen (ER) and progesterone (PR)), human epidermal growth factor-2 (HER2) expression, tumor size, nodal involvement, histologic subtype and grade. Also, using both clinically available information and results from immunohistochemistry and FISH analyses, we will classify cancers into the 'intrinsic' molecular subtypes defined as Luminal A, Luminal B, Basal-like, HER2-expressing and Unclassified. Aim 3) To evaluate the association of MD with each of the histologic and molecular breast cancer subtypes defined in Aim 2, and Aim 4) To combine MD and clinical breast cancer risk factor information to develop a risk prediction model for breast cancer and the specific breast cancer molecular subtypes. Our secondary aim will examine associations of novel parenchyma features of digitized mammogram films with breast cancer and histologic and molecular breast cancer subtypes. Successful completion of this protocol will 1) address the question whether MD is associated with breast cancer subtypes, 2) result in a new risk prediction model for breast cancer and molecular subtypes, and 3) establish a large, collaborative nested-case control study with risk factors, MD, and well-annotated breast cancers that can be used for future studies. Identifying whether MD is associated with specific subtypes of breast cancer may result in more targeted prevention and surveillance strategies for women with high MD. Also, risk models for the molecular subtypes incorporating MD and parenchyma measures will likely be more informative than those treating breast cancer as one disease.